Insight, Impact, Innovation
HomeOur WorkCase StudiesHow can I avoid sending duplicate offers to my customers?

How can I avoid sending duplicate offers to my customers?

Advanced de-duping techniques reduce campaign costs

The Business Challenge

One of the largest prepaid cellular service providers in the U.S. had a database of 17MM customer records. All the customer information was stored at the phone/serial number level and not at customer level, which led our client to believe nearly half of the records were duplicates.

The duplication and discrepancy in customer records, was causing their direct marketing efforts to be incredibly inefficient and this was hampering overall marketing ROI.

The Solution

There was an urgent need to improve the ROI for our client’s marketing initiatives. Identifying unique customers and households was of prime importance for performing any analysis.

Our analysts used advanced tools and text-mining techniques such as pattern matching, phonetic matching and approximate string matching to develop an algorithm for de-duplicating records and identifying unique customers and households. The data was then aggregated at a customer level and later at a household level. Following this we set up a marketing data-mart to perform multi-dimensional churn analysis and deliver customer records for marketing campaigns.


Using advanced de-duplication techniques, we reduced the number of unique customers by 40% and number of households by 46%. This considerably decreased the number of customers to be targeted during campaigns, thereby slicing campaign costs and increasing profitability.

The data aggregation further helped in accurately measuring attrition rates and profiling attriting customers to understand their behavior.  These insights allowed our client to identify areas for improvement. Bi-monthly MIS reporting have also helped our client understand market trends and take quick corrective actions.


Large US Telco company (this client has requested anonymity)


Increase marketing efficiency by identifying and removing duplicate records in large customer database


Fractal's advanced text mining and de-duping techniques


Reduced the number of unique customers by 40% and number of households by 46% thus cutting down campaign costs, increasing profitability and enabling better attrition management